The topic of this application is acquired disorders of action (apraxia) that impact performance of everyday activities. We seek to develop new analytic tools for describing and understanding the neuro-cognitive systems implicated in neurological conditions like ideational apraxia, frontal apraxia, and the executive disorder of traumatic brain injury. These conditions pose a significant clinical problem for neurological rehabilitationists and may place a heavy financial burden on families and society. We advocate a cognitive neuropsychological approach to the problem, which exploits the many parallels between speech and action. The first step involves analyzing the "syntactic" organization of action, and the nature of action errors, in natural action contexts. The Action Coding System (ACS), developed by our group, embodies a descriptive theory of action syntax and methods for coding errors and other measures of disorganization. Interpreted from the perspective of the cognitive systems supporting action, a patient's ACS error profile suggests hypotheses regarding which components of the cognitive system are impaired, and which are spared. These hypotheses may then be examined by the standard cognitive case-study method. This proposal seeks to apply and advance this general approach to the study of action disorders along three specific lines. Part 1 develops a Multi-level Action Test (MLAT) which requires the subject to perform familiar everyday tasks in contexts and combinations that vary along a continuum of task demands. This test will be used as a screening instrument for action disorders of the ideational and frontal types, in patients with TBI and stroke. Part 2 adapts the ACs to the task environment of the MLAT, in an effort to. examine how variations in task demands and contextual constraints effect the expression of action disorders of various types. Part 3 details a program of longitudinal case studies, in which a specially designed battery of neuropsychological tests is administered at regular intervals to a selected subset of patients, in an effort to further specify the nature of the underlying impairment to conceptual or executive systems, and the nature of the change over time. Together, the three parts of this proposal aim to advance the diagnosis of acquired action disorders and their theoretical analysis. The long term goal is to establish predictors of long-term outcome and new approaches to treatment for these often disabling conditions.